Clinicopathological Analysis of Pulmonary Sclerosing Hemangioma

GENERAL THORACIC

Clinicopathological Analysis of Pulmonary Sclerosing Hemangioma Akira Iyoda, MD, Kenzo Hiroshima, MD, Mitsutoshi Shiba, MD, Yukiko Haga, MD, Yasumitsu Moriya, MD, Yasuo Sekine, MD, Kiyoshi Shibuya, MD, Toshihiko Iizasa, MD, and Takehiko Fujisawa, MD Departments of Thoracic Surgery and Basic Pathology, Chiba University, Graduate School of Medicine, Department of Thoracic Surgery, Kimitsu Chuo Hospital, Chiba, Japan

Background. Sclerosing hemangiomas of the lung are uncommon tumors and are thought to be benign. However, the histogenesis and clinicopathological features of these tumors have not been elucidated. Methods. We analyzed the clinicopathological features of 26 sclerosing hemangiomas. The immunoreactivity for Ki-67 and p53 of sclerosing hemangiomas was determined and compared with that of pathological stage 1 pulmonary papillary adenocarcinomas. Results. The patients of sclerosing hemangioma were predominantly female. Eighteen patients were detected as a result of routine medical examinations and 15 were nonsmokers. Seven patients underwent tumor enucleation, 10 underwent a wedge resection, and 9 underwent a lobectomy. The mean tumor size was 2.2 cm (range 1 to 5 cm). Pathological findings demonstrated a papillary pattern in 23 cases, sclerotic pattern in 26 cases, hemor-

rhagic pattern in 22 cases and a solid pattern in 25 cases. Twenty-five cases had an excellent prognosis with no evidence of recurrence following surgery. However, 1 patient who had undergone a wedge resection developed a local recurrence and required an additional wedge resection. The Ki-67 labeling index of sclerosing hemangiomas was significantly lower than that of adenocarcinomas, whereas the Ki-67 labeling index of the recurrent case was 0.4%. No significant immunohistochemical staining for p53 was observed in sclerosing hemangioma cases. Conclusions. Sclerosing hemangioma exhibits various histologic findings. Although we experienced one case with a recurrent tumor, sclerosing hemangiomas did not exhibit malignant behavior. (Ann Thorac Surg 2004;78:1928 –31) © 2004 by The Society of Thoracic Surgeons

S

[5, 6]. Mutations of p53 may be necessary for tumor development and are common in neoplasms with mutant p53 expression being a significant prognostic factor in pulmonary carcinomas [4, 7]. In this study, we examined the clinicopathological behavior and proliferative activity of sclerosing hemangioma using Ki-67 labeling index and p53 expression.

clerosing hemançgioma is an uncommon pulmonary neoplasm and was reported by Liebow and Hubbell in 1956 as a benign neoplasm [1]. Although the morphology of sclerosing hemangioma is well described, the clinicopathologic features or treatment outcomes of sclerosing hemangioma have not been fully elucidated. Subsequent reports of multiple sclerosing hemangioma lesions and lymph node metastases [2, 3] suggest that there may well be cases of malignant sclerosing hemangioma. Therefore, the optimal therapeutic approach of this lesion remains unclear. The histologic examination of sclerosing hemangioma may reveal a papillary pattern and one of the main problems is to differentiate sclerosing hemangioma from carcinoma that exhibits a papillary pattern [2]. The behavior of pulmonary adenocarcinoma is related to a high Ki-67 labeling index as well as p53 expression [4]. The Ki-67 nuclear antigen is associated with cell proliferation and is detectable in the nuclei of cycling (G1, S, G2, and M phase) cells but is absent in resting (G0 phase) cells. It is believed that biologically active tumors express high levels of Ki-67 nuclear antigen Accepted for publication May 12, 2004. Address reprint requests to Dr Fujisawa, Department of Thoracic Surgery, Chiba University, Graduate School of Medicine, Japan, 1– 8 –1, Inohana, Chuo-ku, Chiba 260 – 8670, Japan; e-mail: fujisawat@faculty. chiba-u.jp.

© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc

Material and Methods We reviewed the records of all 26 patients with sclerosing hemangioma who underwent pulmonary resections at Chiba University Hospital between March 1973 and January 2003. The pathology slides were reviewed with particular reference to the presence of papillary, sclerotic, solid and hemorrhagic histologic patterns. In the 16 cases of sclerosing hemangioma with available specimens, immunohistochemical staining was performed with an anti Ki-67 antibody (Dako, Glostrup, Denmark). The Ki-67 labeling index, expressed as the percentage of sclerosing hemangioma nuclei exhibiting positive staining, was determined by light microscopy with random counting of at least 1,000 tumor nuclei. Additional immunohistochemical staining was performed using the monoclonal antibody DO-7 (Dako, Glostrup, Denmark) that specifically reacts with both wild type and mutant p53 protein. The 0003-4975/04/$30.00 doi:10.1016/j.athoracsur.2004.05.069

IYODA ET AL PULMONARY SCLEROSING HEMANGIOMA

1929

Fig 1. Histology of sclerosing hemangioma. (A) Papillary pattern (hematoxylin & eosin stain, ⫻20). (B) Sclerotic pattern (hematoxylin & eosin stain, ⫻20). (C) Hemorrphagic pattern (hematoxylin & eosin stain, ⫻20). (D) Solid pattern (hematoxylin & eosin stain, ⫻20).

Ki-67 labeling index and p53 expression levels were compared with 16 pathological stage 1 pulmonary papillary adenocarcinomas resected surgically between 1988 and 1993.

Immunohistochemical Staining Four-micrometer sections were cut from formalin-fixed paraffin-embedded tissues, placed on silanized slides (Dako, Glostrup, Denmark) and pretreated by microwave treatment for 15 minutes in citrate buffer (10 mmol/L pH 6.0) before immunostaining in order to improve the staining pattern. Tissue sections were stained with the following primary antibodies: monoclonal antibody DO-7 (1:1600; Dako), which reacts with wild type and mutant p53 protein; and a monoclonal antibody against the Ki-67 antigen (1:100; Dako). The streptavidin-biotin method (Histofine SAB-PO Kit; Nichirei, Tokyo, Japan) was used for visualization. DO-7 immunoreactivity was classified as positive when the proportion of positively stained cells was greater than 10% of all tumor cells. Ki-67 labeling indices were determined by light microscopy by randomly counting at least 1,000 tumor nuclei and expressing the results as a percentage of positive cells. The Fisher exact test was used to assess differences in immunohistochemical staining for p53 while the MannWhitney U test was used for analyzing Ki-67 labeling indices.

Results The patient group included 25 women and 1 man and had an average age of 46.2 years old (range 17 to 64 years old). The majority of patients (9 patients) were in their

sixth decade. Eighteen cases were detected during a routine medical examination. One patient complained of cough, whereas 7 were incidentally discovered during investigation for back pain (n ⫽ 2), vertigo (n ⫽ 2), and other diseases (n ⫽ 3). Fifteen of 20 patients for which a smoking history was available were nonsmokers. The diagnosis of sclerosing hemangioma was established by percutaneous needle biopsy (n ⫽ 2), transbronchial forceps biopsy (n ⫽ 2), or transbronchial aspiration biopsy (n ⫽ 3). In 9 cases the diagnosis was established during surgery by stamp cytology (n ⫽ 4) and frozen section (n ⫽ 5). In 10 patients the diagnosis of sclerosing hemangioma was made only after review of stained formalin fixed tissue sections. Seven patients underwent tumor enucleation, 10 underwent wedge resection, and 9 underwent lobectomy. Sixteen tumors were located in the right lung (upper lobe [n ⫽ 4], middle lobe [n ⫽ 4], and lower lobe [n ⫽ 8]) and 10 tumors were in the left lung (upper lobe [n ⫽ 2] and lower lobe [n ⫽ 8]). The mean tumor size was 2.2 cm (range 1 to 5 cm). In 23 patients the sclerosing hemangioma lesions were 30 mm or less in diameter. Pathological findings revealed a papillary pattern (Fig 1A) in 23 cases, sclerotic pattern (Fig 1B) in 26 cases, hemorrhagic pattern (Fig 1C) in 23 cases, and solid pattern (Fig 1D) in 25 cases. Twenty patients exhibited all of the four histologic patterns, 5 patients exhibited three histologic patterns, and 1 exhibited two patterns. Twenty-five patients were followed for an average of 61 months (range 1 to 228) and had no recurrence of sclerosing hemangioma. One case, in which wedge resection had been the initial treatment, suffered local recurrence with a predominantly papillary pattern 4 years after her first operation. Her second operation was

GENERAL THORACIC

Ann Thorac Surg 2004;78:1928 –31

GENERAL THORACIC

1930

IYODA ET AL PULMONARY SCLEROSING HEMANGIOMA

Ann Thorac Surg 2004;78:1928 –31

Table 1. Clinical Characteristics of Sclerosing Hemangioma and Papillary Adenocarcinoma

Gender (male:female) Age (mean) Smoking index (mean) SH ⫽ sclerosing hemangioma;

SH

Ad

1:15 45 93.7

1:15 52 62.5

Ad ⫽ adenocarcinoma.

a wedge resection, and she was followed for 10 more years during which time there was no evidence of recurrence. We compared the Ki-67 labeling index and p53 expression of sclerosing hemangioma cases with pathological stage 1 papillary adenocarcinoma. It should be noted that there was no significant difference in age, gender and smoking index between these two groups (Table 1). The mean Ki-67 labeling index of sclerosing hemangioma was significantly lower than that of adenocarcinomas (p ⬍ 0.0001; Table 2), although the Ki-67 labeling index of recurrent cases was 0.4%. There was no immunohistochemical staining for p53 evident in the sclerosing hemangioma cases (p ⫽ 0.0068; Table 2).

Comment Sclerosing hemangioma is a rare tumor and a detailed clinicopathological study for sclerosing hemangioma is merited. It is noteworthy that more than 95% of patients with sclerosing hemangioma were women. This gender difference is more pronounced in our study than in the study of Devouassoux-Shisheboran and colleagues [8], although the mean age of the patients in these two studies was comparable. Sclerosing hemangioma is histologically composed of a mixture of papillary, solid, sclerotic, or hemorrphagic patterns. Most sclerosing hemangioma cases exhibit a papillary pattern [9]. In a study of 51 patients by Katzenstein and coworkers [2], the solid pattern was present in all patients, the hemorrhagic pattern in 37, the papillary pattern in 38, and the sclerotic pattern was evident in 50 patients. They indicated that diagnoses of metastatic papillary thyroid carcinoma, mesothelioma, and especially bronchioloalveolar carcinoma might well be considered if the papillary pattern predominated. Sugio and associates [9] demonstrated that the solid pattern was present in 8 of 10 patients, the papillary pattern in 5, the sclerotic pattern in all, and the hemangiomatous pattern in 9 patients. In our study, the majority of cases exhibited

Table 2. Ki-67 Labeling Index and p53 Expression in Sclerosing Hemangioma and Adenocarcinoma

Ki-67 labeling index p53 positivity

SH

Ad

p Value

0.19 ⫾ 0.18% 0/16 (0%)

9.25% ⫾ 5.55% 7/16 (43.8%)

⬍ 0.0001 0.0068

SH ⫽ sclerosing hemangioma;

Ad ⫽ adenocarcinoma.

the typical four patterns with the sclerotic pattern being revealed in all cases. The histogenesis and clinical behavior of sclerosing hemangioma are unknown. Previously, it has been proposed that pulmonary sclerosing hemangioma cells originate from pulmonary epithelial [10, 11], mesothelial [12], mesenchymal [13] or endothelial cells [14]. However, in the largest series reported to date by Devouassoux-Shisheboran and colleagues [8], clinicopathologic study of 100 pulmonary sclerosing hemangioma patients suggested that sclerosing hemangioma cells were derived from primitive respiratory epithelium. The basis for this suggestion was their finding of TTF-1 expression in round cells of sclerosing hemangioma in the absence of concomitant expression of surfactant proteins A and B. Although sclerosing hemangioma is thought to be a benign lung tumor, there are reported cases that exhibit multiple nodules or lymph node metastases [2, 3]. Katzenstein and coworkers [2] reported two patients (5.1%) with multiple sclerosing hemangioma lesions. Devouassoux-Shisheboran and associates [8] reported that one 3.5-cm peripheral sclerosing hemangioma was associated with involvement of multiple peribronchial lymph nodes. In our study, there was no case with lymph node metastasis or multiple nodules. However, we did experience one case with recurrent tumor [15]. Therefore, we performed immunohistochemical examination for Ki-67 and p53 expression in order to assess the biological activity of sclerosing hemangioma. The Ki-67 labeling index and p53 expression were significantly lower in sclerosing hemangioma than in stage 1 papillary adenocarcinoma. These results indicate that the sclerosing hemangioma patients do not exhibit a ‘malignant phenotype’. Furthermore, the Ki-67 labeling index of the recurrent tumor in our only case exhibiting recurrent disease was low. This indicates a low proliferative activity in this recurrent case and therefore suggests that this recurrence might have been secondary to incomplete resection of the tumor. It is important that the diagnosis of sclerosing hemangioma be established preoperatively whenever possible because a limited, but complete, resection of the lesion is the treatment of choice [9]. However, in some cases it may be difficult to establish a definitive diagnosis intraoperatively, particularly in sclerosing hemangioma cases exhibiting a predominant papillary pattern or multiple nodules [2]. The cytologic features of sclerosing hemangioma are still relatively unclear as there have been few studies in this area. We are aware of 3 case reports [16 –18], including that of Wang and colleagues [16] in which it was difficult to differentiate sclerosing hemangioma from bronchioloalveolar carcinoma or metastatic papillary adenocarcinoma. However, our previous studies [19, 20] indicated that sclerosing hemangioma included a variety of epithelial and nonepithelial cells and that there were significant differences between epithelial cells of sclerosing hemangioma and papillary adenocarcinomas. Therefore, we need to perform further studies

to address the preoperative diagnosis of sclerosing hemangioma. In this study, we experienced one case with rare recurrent tumor. However, study of proliferative activity and p53 status of sclerosing hemangiomas did not provide any evidence of a malignant phenotype. Because sclerosing hemangioma is a tumor with unknown histogenesis and unclear biological behavior, we are still required to take great care during surgical removal and follow-up in patients with these tumors. This work was supported in part by a Grant-in-Aid for Scientific Research (C)(2) 15591467 of the Japanese Ministry of Education, Culture, Sports, Science and Technology.

IYODA ET AL PULMONARY SCLEROSING HEMANGIOMA

9. 10.

11.

12. 13.

References 1. Liebow AA, Hubbell DS. Sclerosing hemangioma (histiocytoma, xanthoma) of the lung. Cancer 1956;9:53–75. 2. Katzenstein ALA, Gmelich JT, Carrington CB. Sclerosing hemangioma of the lung. A clinicopathologic study of 51 cases. Am J Surg Pathol 1980;4:343–56. 3. Noguchi M, Kodama T, Morinaga S, Shimosato Y, Saito T, Tsuboi E. Multiple sclerosing hemangiomas of the lung. Am J Surg Pathol 1986;10:429 –35. 4. Haga Y, Hiroshima K, Iyoda A, et al. Ki-67 expression and prognosis for smokers with resected stage 1 non-small cell lung cancer. Ann Thorac Surg 2003;75:1727–33. 5. Gerdes J, Lemke H, Baisch H, Wacker H, Schwab U, Stein H. Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 1984;133:1710 –5. 6. Cattoretti G, Becker MH, Key G, et al. Monoclonal antibodies against recombinant parts of the Ki-67 antigen (MIB-1 and 3) detect proliferating cells in microwave-processed formalin-fixed paraffin sections. J Pathol 1992;168:357– 63. 7. Lee YC, Chang YL, Luh SP, Lee JM, Chen JS. Significance of p53 and Rb protein expression in surgically treated nonsmall cell lung cancers. Ann Thorac Surg 1999;68:343– 8. 8. Devouassoux-Shisheboran M, Hayashi T, Linnoila RI, Koss MN, Travis WD. A clinicopathologic study of 100 cases of pulmonary sclerosing hemangioma with immunohisto-

14. 15.

16. 17. 18. 19.

20.

1931

chemical studies. TTF-1 is expressed in both round and surface cells, suggesting an origin from primitive respiratory epithelium. Am J Surg Pathol 2000;24:906 –16. Sugio K, Yokoyama H, Kaneko S, Ishida T, Sugimachi K. Sclerosing hemangioma of the lung: radiographic and pathological study. Ann Thorac Surg 1992;53:295–300. Nagata N, Dairaku M, Ishida T, Sueishi K, Tanaka K. Sclerosing hemangioma of the lung. Immunohistochemical characterization of its origin as related to surfactant apoprotein. Cancer 1985;55:116 –23. Satoh Y, Tsuchiya E, Weng SY, et al. Pulmonary sclerosing hemangioma of the lung. A type II pneumocytoma by immunohistochemical and immunoelectron microscopic studies. Cancer 1989;64:1310 –7. Katzenstein ALA, Fulling K, Weise D, Battifora H. So-called sclerosing hemangioma of the lung. Evidence for mesothelial origin. Am J Surg Pathol 1983;7:3–14. Hsu SM, Raine L, Fanger H. The use of avidin-biotinperoxidase complex in immunoperoxidase techniques. Am J Clin Pathol 1981;75:816 –21. Kay S, Still WJ, Borochovitz D. Sclerosing hemangioma of the lung: an endothelial or epithelial neoplasm? Hum Pathol 1977;8:468 –74. Odaka E, Shiba M, Mitsunaga S, Yamakawa H, Ogawa T, Yamaguchi Y. A resected case of recurrent pulmonary sclerosing hemangioma. Haigann 1993;33:281–7 (in Japanese with English abstract). Wang SE, Nieberg RK. Fine needle aspiration cytology of sclerosing hemangioma of the lung, a mimicker of bronchioloalveolar carcinoma. Acta Cytol 1986;30:51– 4. Chow LTC, Chan SK, Chow WH, Tsui MS. Pulmonary sclerosing hemangioma. report of a case with diagnosis by fine needle aspiration. Acta Cytol 1992;36:287–92. Kaw YT, Nayak RN. Fine needle aspiration biopsy cytology of sclerosing hemangioma of the lung. A case report. Acta Cytol 1993;37:933–7. Shiba M, Odaka E, Mitsunaga S, Yamakawa H, Yamaguchi Y. A clinicopathological study of resected sclerosing hemangioma of the lung - its preoperative cytodiagnosis. J Jpn Soc Clin Cytol 1992;31:378 – 85 (in Japanese with English abstract). Iyoda A, Baba M, Saitoh H, et al. Imprint cytologic features of pulmonary sclerosing hemangioma. Comparison with well-differentiated papillary adenocarcinoma. Cancer 2002; 96:146 –9.

GENERAL THORACIC

Ann Thorac Surg 2004;78:1928 –31